I agree (surprise!)....people switch to WAL mode for concurrent read/write. I would've assumed a write rollback was non-intrusive into exisiting reads in all circumstances unless told otherwise. So some sort of cross-reference to shared cache causing potential problems under rollbacks would be more than appreciated.
Michael D. Black Senior Scientist Advanced Analytics Directorate Advanced GEOINT Solutions Operating Unit Northrop Grumman Information Systems ________________________________ From: sqlite-users-boun...@sqlite.org [sqlite-users-boun...@sqlite.org] on behalf of Brad House [b...@monetra.com] Sent: Tuesday, May 01, 2012 8:22 AM To: General Discussion of SQLite Database Subject: EXT :Re: [sqlite] SQLite 3.7.11 behavioral change from 3.7.10 with concurrency -- bug?? -- test case! I agree that the docs for shared-cache mode need to be updated to reflect the disadvantage which causes unexpected (non-standard) behavior, and also emphasize that you probably should not be enabling the feature unless you are on a memory-limited device due to these behavioral differences. Thanks. -Brad _______________________________________________ _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
